# Deformable Video Transformer

arXiv: https://arxiv.org/abs/2203.16795 · HTML: https://mlsanity.com/paper/2203.16795/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 40 of its 46 references that are in the map.
- First posted on arXiv 31 Mar 2022.

## Builds on (most cited first, 30 of 40)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [Big Bird: Transformers for Longer Sequences](https://mlsanity.com/paper/2007.14062/) (2020)
- [Reformer: The Efficient Transformer](https://mlsanity.com/paper/2001.04451/) (2020)
- [Is Space-Time Attention All You Need for Video Understanding?](https://mlsanity.com/paper/2102.05095/) (2021)
- [ViViT: A Video Vision Transformer](https://mlsanity.com/paper/2103.15691/) (2021)
- [MLP-Mixer: An all-MLP Architecture for Vision](https://mlsanity.com/paper/2105.01601/) (2021)
- [A Closer Look at Spatiotemporal Convolutions for Action Recognition](https://mlsanity.com/paper/1711.11248/) (2017)
- [The "something something" video database for learning and evaluating visual common sense](https://mlsanity.com/paper/1706.04261/) (2017)
- [Video Swin Transformer](https://mlsanity.com/paper/2106.13230/) (2021)
- [Two-Stream Convolutional Networks for Action Recognition in Videos](https://mlsanity.com/paper/1406.2199/) (2014)
- [Temporal Segment Networks: Towards Good Practices for Deep Action Recognition](https://mlsanity.com/paper/1608.00859/) (2016)
- [Multiscale Vision Transformers](https://mlsanity.com/paper/2104.11227/) (2021)
- [TSM: Temporal Shift Module for Efficient Video Understanding](https://mlsanity.com/paper/1811.08383/) (2018)
- [Deformable ConvNets v2: More Deformable, Better Results](https://mlsanity.com/paper/1811.11168/) (2018)
- [X3D: Expanding Architectures for Efficient Video Recognition](https://mlsanity.com/paper/2004.04730/) (2020)
- [CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows](https://mlsanity.com/paper/2107.00652/) (2021)
- [Rethinking Spatiotemporal Feature Learning: Speed-Accuracy Trade-offs in Video Classification](https://mlsanity.com/paper/1712.04851/) (2017)
- [Efficient Content-Based Sparse Attention with Routing Transformers](https://mlsanity.com/paper/2003.05997/) (2020)
- [VATT: Transformers for Multimodal Self-Supervised Learning from Raw Video, Audio and Text](https://mlsanity.com/paper/2104.11178/) (2021)
- [Convolutional Two-Stream Network Fusion for Video Action Recognition](https://mlsanity.com/paper/1604.06573/) (2016)
- [Video Transformer Network](https://mlsanity.com/paper/2102.00719/) (2021)

## Built on by (most cited first, 15 of 15)

- [Selective Structured State-Spaces for Long-Form Video Understanding](https://mlsanity.com/paper/2303.14526/) (2023)
- [In the Eye of Transformer: Global-Local Correlation for Egocentric Gaze Estimation](https://mlsanity.com/paper/2208.04464/) (2022)
- [Motion-Guided Masking for Spatiotemporal Representation Learning](https://mlsanity.com/paper/2308.12962/) (2023)
- [Vision Transformers for Action Recognition: A Survey](https://mlsanity.com/paper/2209.05700/) (2022)
- [Video Token Merging for Long-form Video Understanding](https://mlsanity.com/paper/2410.23782/) (2024)
- [Linear Video Transformer with Feature Fixation](https://mlsanity.com/paper/2210.08164/) (2022)
- [Enhancing Action Recognition by Leveraging the Hierarchical Structure of Actions and Textual Context](https://mlsanity.com/paper/2410.21275/) (2024)
- [SpikeVideoFormer: An Efficient Spike-Driven Video Transformer with Hamming Attention and O(T) Complexity](https://mlsanity.com/paper/2505.10352/) (2025)
- [A Simple Video Segmenter by Tracking Objects Along Axial Trajectories](https://mlsanity.com/paper/2311.18537/) (2023)
- [PosMLP-Video: Spatial and Temporal Relative Position Encoding for Efficient Video Recognition](https://mlsanity.com/paper/2407.02934/) (2024)
- [Towards Robust Video Instance Segmentation with Temporal-Aware Transformer](https://mlsanity.com/paper/2301.09416/) (2023)
- [SVT: Supertoken Video Transformer for Efficient Video Understanding](https://mlsanity.com/paper/2304.00325/) (2023)
- [Towards Real-World Video Denosing: A Practical Video Denosing Dataset and Network](https://mlsanity.com/paper/2207.01356/) (2022)
- [PatchBlender: A Motion Prior for Video Transformers](https://mlsanity.com/paper/2211.14449/) (2022)
- [Semantic-Aware Local-Global Vision Transformer](https://mlsanity.com/paper/2211.14705/) (2022)

## Abstract (arXiv)

Video transformers have recently emerged as an effective alternative to convolutional networks for action classification. However, most prior video transformers adopt either global space-time attention or hand-defined strategies to compare patches within and across frames. These fixed attention schemes not only have high computational cost but, by comparing patches at predetermined locations, they neglect the motion dynamics in the video. In this paper, we introduce the Deformable Video Transformer (DVT), which dynamically predicts a small subset of video patches to attend for each query location based on motion information, thus allowing the model to decide where to look in the video based on correspondences across frames. Crucially, these motion-based correspondences are obtained at zero-cost from information stored in the compressed format of the video. Our deformable attention mechanism is optimised directly with respect to classification performance, thus eliminating the need for suboptimal hand-design of attention strategies. Experiments on four large-scale video benchmarks (Kinetics-400, Something-Something-V2, EPIC-KITCHENS and Diving-48) demonstrate that, compared to existing video transformers, our model achieves higher accuracy at the same or lower computational cost, and it attains state-of-the-art results on these four datasets.